Neural Networks
Weights, loss and backprop — how a model turns examples into rules nobody wrote by hand.
AI Master.
Everything I'm learning about working with AI and data, filed in one place. Open the folder, pull a page out, or click one to read it up close.
Data Pipelines
Clean, label and shape raw data before a model ever sees it. Garbage in, garbage out.
LLMs & Prompting
Tokens, embeddings and context windows — prompts written with intent, not luck.
RAG & Vectors
Ground answers in your own data: embed it, index it, retrieve the right chunk, cite it.
Orchestration
An orchestrator splits the job and hands it to specialist agents that call tools over MCP.
Evals & Guardrails
Measure accuracy, catch hallucinations and ship with guardrails — AI you can trust in prod.
▶ click to open ✋ drag a page out 🔍 click a page to zoom